On the left bank of this supposed river only one habitation, opposite No. 1, was discovered, which M. Vouga thought had already been pillaged. It was reported that near this spot several human skeletons had been discovered, one of which had a rope round the neck! Below this the channel becomes deeply buried, and the superficial gravel attains the depth of some ten feet; but, nevertheless, Vouga succeeded in making excavations which decided the chronological sequence of the Roman and Gallic remains. "It was," says he, "in the midst of these gravels that I found the layer containing Roman remains--tiles and fragments of pottery, nails, etc.--at a height of two and a half to three feet above the Gallic objects." (B. 428, p. 13.) These Gallic objects consisted of the well-known fibulæ and other articles characteristic of La Tène, so that superposition clearly indicates the Roman occupation to be posterior to that of its original constructors. M. Vouga believes that the channel, along the banks of which he found the remains of so many houses, was the right branch of two outlets which at that time existed, and which united lower down to form the Thielle. The left branch was nearer the rising ground towards Epagnier, but it is now covered over with gravel, and has never yet been examined. Some 300 or 400 yards lower down there are some gravel pits, which are occasionally worked for road metal, in which I saw in the summer of 1886 a great many piles, singly and in groups, cropping up through a black peaty deposit underneath the gravel. I mentioned the matter to M. Vouga, and he informed me that the few things found there indicate a Gallo-Roman period. According to M. Vouga, the site of La Tène station extended from the south bank of the outlet to the small island formed by its two branches before they became united. The upper part of this island, now denuded of its peaty deposits by the action of the waves, forms part of the lake. This denuding process is still going on at the margin of the lake all the way from La Tène to Préfargier. Large masses of the ancient sedimentary deposits, containing piles and relics of the lake-dwellers, become undermined and broken up by the waves, leaving their more solid relics, such as stone hatchets, mixed with the gravel. These are often thrown up on the beach, and in this way many beautiful jade hatchets have been picked up from the sites of the four lake-dwellings now almost entirely disintegrated, which existed along the north shore from La Tène to Préfargier. It is in the gravel at the upper end of La Tène that most of the coins have been collected.